Whisper Enricher#
Module type
Integrates with a Whisper API service to transcribe, translate, or detect the language of audio and video files.
Features#
Submits audio or video files to a Whisper API deployment for processing.
Supports operations such as transcription, translation, and language detection.
Optionally generates SRT subtitle files for video content.
Integrates with S3-compatible storage systems to make files publicly accessible for processing.
Handles job submission, status checking, artifact retrieval, and cleanup.
Notes#
Requires a Whisper API endpoint and API key for authentication.
Only compatible with S3-compatible storage systems for media file accessibility.
** This stores the media files in S3 prior to enriching them as Whisper requires public URLs to access the media files.
Handles multiple jobs and retries for failed or incomplete processing.
Configuration Options#
YAML#
whisper_enricher:
api_endpoint: ''
api_key: ''
include_srt: false
timeout: 90
action: translate
Command Line:#
Option |
Description |
Default |
Type |
|---|---|---|---|
|
Required. WhisperApi api endpoint, eg: https://whisperbox-api.com/api/v1, a deployment of bellingcat/whisperbox-transcribe. |
string |
|
|
Required. WhisperApi api key for authentication |
string |
|
|
Optional. Whether to include a subtitle SRT (SubRip Subtitle file) for the video (can be used in video players). |
False |
string |
|
Optional. How many seconds to wait at most for a successful job completion. |
90 |
string |
|
Optional. which Whisper operation to execute |
translate |
string |